Polyolefins have many outstanding characteristics, and are widely used in many applications. However, their yield strength and stiffness are relatively poor compared to other engineering materials. These properties are commonly enhanced in other polymers by forming a composite with high strength fibers. The outstanding specific strength and stiffness of carbon fibers has led to the wide use of carbon fiber-polymer matrix composites across a wide range of applications. However, polyolefins exhibit very weak interactions with carbon materials because they are free from polar groups and chemically inactive. Since melted polyolefins do not readily wet carbon fibers, voids can form in the composite during production. Additionally, very high molecular weight polyolefins, such as ultrahigh molecular weight polyethylene (UHMWPE) have very high viscosity melt states and do not readily impregnate the fibers even when melted. Moreover, the finished composite exhibits weak interactions between the carbon fibers and the polyolefin matrix.
One method for forming continuous fiber composites with polyolefins begins by making an aqueous slurry of the polyolefin powder and passing the fibers through the slurry (Tang et al., 1997). Alternately, the slurry can be impregnated into a web of fibers, as demonstrated for UHMWPE powder and carbon fibers (US 2005/0287891). The UHMWPE had a molecular weight (by convention the number average molecular weight) of 2.5×106 amu, which is a relatively low value for UHMWPE, and this lower molecular weight UHMWPE processes easier than the UHMWPE with a number average molecular weight of about 5×106 amu, which is desired for use in wear-resistant parts. The water is then removed by drying before pressing the composite article. This technique does not overcome the problem of poor fiber-matrix interaction, as illustrated by the loss of impact strength after carbon fiber addition relative to before carbon fiber addition in the slurry-impregnated example of US 2005/0287891 (see paragraph [0032], in US 2005/0287891).
High adhesion between carbon fibers and a thermoplastic resin (matrix resin) is needed for improved mechanical properties. Methods for increasing wettability, and therefore increasing the polymer-fiber adhesion, by functionalizing the surface of carbon fibers through plasma treatment, ozone treatment, or chemical etching have been used, but unfortunately these methods damage the carbon fibers. Also, many methods have been proposed for producing a modified resin by graft polymerizing a polymerizable monomer containing a polar functional group onto an olefin resin. For example, US 2012/0295504 teaches that carbon fiber-polyolefin composites may be made by modifying the polyolefin to include a monomer containing an ethylenic double bond, such as styrene, and then stacking alternate layers of carbon fiber cloth (or unitape) and the modified polyolefin/polyolefin blend and subsequently hot pressing. In this case, the modifications to the polyolefin alter the properties of the matrix in a potentially undesirable manner. For example, the wear resistance of UHMWPE is compromised when other polymers are grafted onto it, or it is blended with other polymers. Also, as described below, UHMWPE does not penetrate the carbon fiber cloth during hot pressing.
Another means of forming carbon fiber composites with mixed polyolefin and polyamide matrices is to pull or extrude continuous carbon fibers through a melted mixture of polyolefin and polyamide, then chopping the fiber/polymer blends transversely to the length of the fiber and compounding them into the final product. This technique, given in U.S. Pat. No. 5,409,763 and U.S. Pat. No. 6,794,032 for example, is not applicable to UHMWPE because the melt viscosity of UHMWPE is too high to be used in that manufacturing technique. Furthermore, since only a single fiber bundle is passed through the die at a time, assembling the coated fibers into a ply would require additional weaving after coating, which can damage the polymer coating due to the low hardness of most polymers.
U.S. Pat. No. 8,877,834 teaches that a melted blend of polyolefin and acid-modified polyolefin can be used to make composites that have good load transfer with carbon fibers previously coated with an amine-containing size. Again, the melt state of UHMWPE has too high of a viscosity to coat fibers during extrusion through a die. Furthermore, the addition of the acid-modified polyolefin may detrimentally affect the properties of the composite in comparison to the pure polyolefin, for example the wear resistance.
U.S. Pat. No. 4,804,427 teaches coating continuous carbon (or other) fibers with an epoxy resin and initiator, curing the epoxy, and then polymerizing a thermoplastic (including polybutadiene) in the presence of the cured epoxy-coated carbon fiber. Coating the fibers with uncured resin and then curing the resin significantly complicates processing and increases the composite cost. Furthermore, neither polyethylene nor polypropylene were contemplated, which are the most difficult polyolefin matrices to achieve load transfer to the fibers.
A few publications describe a method for anchoring catalysts on certain types of fibers, and polymerizing olefins. For example, Etcheverry and Barbosa (2012) treated relatively long glass fibers with methylalumoxane (MAO, a co-catalyst for polyolefin production from metallocene catalysts), and polymerized propylene with the treated fibers in a solvent. In this technique, the MAO requires hydroxyl groups on the fiber surface. While hydroxyl groups dominate glass fiber surfaces, they are largely not present on carbon fiber surfaces. Additionally, only a very thin coating of PP was obtained on the fibers. A similar catalytic grafting technique was also used to produce PE-coated asbestos, UHMWPE and Kevlar® fibers (Wang et al., 1992, 1992, and 1993). DuPont™ Kevlar® is an aromatic polyamide fiber. In this method, titanium chloride was decomposed on the fiber surface to make a Ziegler-type catalyst. The success of the polymerization varied according to the number of hydroxyl groups on the fiber surface; the UHMWPE and Kevlar® fibers were plasma-treated in order to introduce the necessary hydroxyl groups. The Kevlar® and asbestos fibers were short fibers (length not given, though), free to circulate through the solvent during catalyst deposition and polymerization. The UHMWPE monofilaments were tightly wound on high density polyethylene (HDPE) frames and fitted together in the reactor. While experimental detail is lacking, it appears as if the fibers were not removed from the frames before hot pressing with additional HDPE to form the composites. These methods have the disadvantage of requiring catalyst preparation for each polymerization. Furthermore, the metallic catalyst (and presumably co-catalyst) is retained in the composite, since it was covalently bonded to the fiber surface through the hydroxyl groups. Metal content in carbon fiber-polymer matrix composites would prevent their use in certain applications, including implanted biomaterials and ablative insulation for military reentry vehicles.
Olefins may be polymerized (in-situ) in the presence of very short, discontinuous high aspect ratio carbons, primarily carbon nanotubes and nanofibers. This is usually accomplished by a homogeneous molecular catalyst, and polyethylene is the most typical product (Bonduel et al., 2005; Park and Choi, 2009). Non-molecular Ziegler-Natta catalysts have also been deposited on the nanotubes for polyethylene growth (Amoli et al., 2011). Polypropylene (Funck and Kaminsky, 2007) nanocomposites have also been formed by this technique. The discontinuous reinforcements are isotropically distributed throughout the reaction vessel during polymerization, and rapidly moving as the solvent is stirred. This method is not applicable to continuous carbon fibers, which cannot be isotropically distributed in the reaction vessel, are essentially immobile compared to the solvent and its dissolved contents, and would rapidly wrap around a stirring shaft and/or become balled up during agitation.
Similarly, composites of UHMWPE with chopped carbon fibers have long been known, and were briefly marketed in the late 1970s as materials for wear surfaces in hip implants (reviewed in Kurtz, 2009). The material was known as Poly II, but its performance was quite poor due to a complete lack of binding between the matrix and the fibers. As the matrix was worn, the chopped fibers simply pulled out from the composite. Other work with discontinuous carbon fiber (chopped or vapor-grown nanofibers) reinforcements in polyethylene and polypropylene composites prepared by typical polymer compounding techniques (twin-screw extrusion, melt blending, etc.) was recently reviewed (Shaikh et al., 2014). All of these composites were formed by mechanical mixing of the UHMWPE with the chopped fibers, a method that is not applicable to continuous fibers.
These references contain at least one of the following limitations in regard to fiber-polyolefin composites, in particular for fiber-polyolefin composites used as a wear surface for medical implants: loss of impact strength of the composite relative to the pure polyolefin; poor polymer adhesion to the fibers; only short and discontinuous fibers can be used in the production methods; fibers are damaged during the composite preparation (for example etching, plasma, and the like); the olefin is chemically altered or blended with a second material and this leads to degradation of a desired mechanical property such as wear resistance; there is poor intermixing of polyolefin polymer chains and the fibers which leads to poor mechanical properties of the composite; or there is residual catalyst or metal content which is unacceptable for high purity applications (for example medical implants).
As such, there remains a need for a continuous fiber-reinforced polyolefin composite which has: an improved impact resistance for the composite relative to the pure polyolefin, a high wear resistance (in particular for use in medical implants), a high purity (having little residual metals left from the polymerization catalysts used to produce the polyolefins), long, continuous fibers and a high molecular weight polyolefin intimately intermixed with the long fibers that lack chemically bound polymerization catalysts, all which overcome the limitations of the prior art.